water resources and waste treatment, monitoring and controlling water and
providing real-time information to help water companies and households manage
their water better. Smart water networks have been described as layered
architecture, beginning with the sensing-and-control layer through data collection
and data management and ending with the data fusion-and-analytics layer [27].
Although the technology components for smart water cities are available, the route
to application is uncertain. The main hurdles are lack of integrated and open
solutions, difficulty to comply with user and integration requirements, lack of
clear and validated business cases for solutions, lack of business intelligence
awareness and lack of political and regulatory support.
The quantity and complexity of sensor and environmental data is growing at
an increasing rate, while the demands for new solutions and tools to utilise and
interpret this data are likewise growing due to financial and regulatory pressures.
The phrase ‘big data’ may then become a reality for the water sector particularly
on the customer side, since when smart metering becomes more prevalent a
huge amount of data will be collected. If the UK goes to a point where the
entire water industry is universally metered with smart metering, there will be
approximately 25 million water meters for customers. Organising, managing
and supporting such massive ICT network infrastructure, however, are substantial
technical challenges. This data could be used, in conjunction with mapping software
and hydraulic models to map consumption in DMAs where there are spikes in usage.
As demand for clean water increases with population growth in the coming
decades and supply remains stagnant or shrinks due to climate change, solutions to
manage and minimise leaks will become increasingly critical. Many water utilities
are struggling to measure and locate leaks in their distribution networks beyond
the economic level of leakage, and there is a drive to efficiency by implementing
leak-reducing solutions. Leakage results in wasted energy costs (such as spent
pumping water), water treatment costs (energy and chemicals), misdirected repair
activities, regulatory penalization and environmental damage to city infrastructure.
Smart water networks offer the potential to identify leaks early, thus reducing the
amount of water that is wasted and saving utilities money. These solutions include
the use of flow and pressure sensors to gather data, analyse the data using algorithms
to detect patterns that could reveal a leak in the network and provide real-time data
on the location of a leak. A condition monitoring approach for smart networks
can allow the early detection of potential faults in assets. The integration of realtime analytics can facilitate rapid determination (i.e. before customers are impacted)
of abnormal flow events.
A number of approaches from the fields of artificial intelligence and statistics
have been applied for detecting abnormality in WDSs from time series data.
Alert systems that convert flow and pressure sensor data into usable information
in the form of timely alerts (event detection systems) have been developed with a
focus on burst detection to help with the issue of leakage reduction. Analysis systems
need to provide useful classifications of system status, events and conditions
and not provide an onerous number of alerts or alarms to system operators who
will otherwise ignore warnings hence compromising the value of the information.
Data Science Trends and Opportunities for Smart Water Utilities
17
providing real-time information to help water companies and households manage
their water better. Smart water networks have been described as layered
architecture, beginning with the sensing-and-control layer through data collection
and data management and ending with the data fusion-and-analytics layer [27].
Although the technology components for smart water cities are available, the route
to application is uncertain. The main hurdles are lack of integrated and open
solutions, difficulty to comply with user and integration requirements, lack of
clear and validated business cases for solutions, lack of business intelligence
awareness and lack of political and regulatory support.
The quantity and complexity of sensor and environmental data is growing at
an increasing rate, while the demands for new solutions and tools to utilise and
interpret this data are likewise growing due to financial and regulatory pressures.
The phrase ‘big data’ may then become a reality for the water sector particularly
on the customer side, since when smart metering becomes more prevalent a
huge amount of data will be collected. If the UK goes to a point where the
entire water industry is universally metered with smart metering, there will be
approximately 25 million water meters for customers. Organising, managing
and supporting such massive ICT network infrastructure, however, are substantial
technical challenges. This data could be used, in conjunction with mapping software
and hydraulic models to map consumption in DMAs where there are spikes in usage.
As demand for clean water increases with population growth in the coming
decades and supply remains stagnant or shrinks due to climate change, solutions to
manage and minimise leaks will become increasingly critical. Many water utilities
are struggling to measure and locate leaks in their distribution networks beyond
the economic level of leakage, and there is a drive to efficiency by implementing
leak-reducing solutions. Leakage results in wasted energy costs (such as spent
pumping water), water treatment costs (energy and chemicals), misdirected repair
activities, regulatory penalization and environmental damage to city infrastructure.
Smart water networks offer the potential to identify leaks early, thus reducing the
amount of water that is wasted and saving utilities money. These solutions include
the use of flow and pressure sensors to gather data, analyse the data using algorithms
to detect patterns that could reveal a leak in the network and provide real-time data
on the location of a leak. A condition monitoring approach for smart networks
can allow the early detection of potential faults in assets. The integration of realtime analytics can facilitate rapid determination (i.e. before customers are impacted)
of abnormal flow events.
A number of approaches from the fields of artificial intelligence and statistics
have been applied for detecting abnormality in WDSs from time series data.
Alert systems that convert flow and pressure sensor data into usable information
in the form of timely alerts (event detection systems) have been developed with a
focus on burst detection to help with the issue of leakage reduction. Analysis systems
need to provide useful classifications of system status, events and conditions
and not provide an onerous number of alerts or alarms to system operators who
will otherwise ignore warnings hence compromising the value of the information.
Data Science Trends and Opportunities for Smart Water Utilities
17